Informal caregivers are a vital part of the health care system and their importance will grow with the aging of the U.S. population. Research has found a significantly higher risk of death among older caregivers who reported emotional strain due to caregiving. However, the pathway to this increased mortality risk is not clear. Studies have shown an association between caregiving and preclinical and clinical disease, but there is a large gap in scientific knowledge about the effect of caregiving on morbidity and mortality. Our general hypothesis is that both self-care and formal care is compromised among caregivers, especially those experiencing emotional strain, resulting in poorer medical management, less continuity of care, and greater progression of the caregiver's treatable conditions, as shown by patterns of health service use. Currently available evidence concerning this hypothesis is limited and results are contradictory. Most studies: 1) treated caregiving as a static construct, rather than a life- cycle construct (with before, during and after caregiving stages), 2) used clinical or volunteer samples which may not be representative of all caregivers and 3) which have difficulty identifying reliably comparable non-caregivers, and 4) identified health care service retrospectively by self-report which limits precision and detail of information. We propose to examine the effect of caregiving on health by investigating differences in health care service use across the caregiving life cycle in a biracial population-based sample of 599 people age 65 or older. To uniformly assess health service utilization, we will use Medicare billing data over the specific interval during which caregiving stage and other relevant characteristics were identified. Because data are already available from a longitudinal, population-based study of caregivers, the hypotheses can be tested at small cost. We hypothesize that during caregiving, people will have fewer preventive and routine physician visits, less continuity of care, and physician visits with higher intensity of care than non-caregivers or former caregivers. During and after caregiving people will have more emergency department (ED) visits, more urgent and emergent ED visits, more hospitalizations, and more hospitalizations for ambulatory care sensitive conditions (ACSC) than non-caregivers. We further hypothesize that these relationships will be similar for non-Hispanic blacks and whites and that among caregivers, service use will be modified by level of emotional strain. PUBLIC HEALTH RELEVANCE: If results of this work suggest caregivers neglect their own health it could point to ways of reducing the excess mortality associated with caregiving. Most interventions focus on the needs of the care recipient;this work may suggest ways caregivers should be assisted to care for their own health.